gay blue dog
i retain a pretty dismal view of AI for just about any use case, but had some distant friends / people i follow on social media say they used it as a rubber duck for troubleshooting a problem they had, or a place to just dump emotions into. i figured this, at the very minimum, could and should be harmless. i guess i wasn’t cynical enough
your third sentence here is a non-sequitur – do you mean to say disposable razors better work on longer hair that safety razors?
i aim to please
alright, fine, i’ll do it.
webshit weekly (2025/03/27)
How to Use Em Dashes (—), En Dashes (–) , and Hyphens (-)
Grammar Nazis (as opposed to the regular kind) publish a guide on how to best calibrate your printing press to 17th-century standards. Several Hackernews (some of which are the regular kind) offer their own competing, more-detailed guides in response. The concern is raised that using too many typographic dashes makes you sound like ChatGPT, to much dismay of those still diligently copying from the Google (business model: “Uber for glue pizza”) results page for “em dash”. Multiple Hackernews take the opportunity call the group of people who do not care about the millimeter difference between the types of dashes “NPCs”.
claiming to have customers you don’t actually have so vocally that they have to sue you to get their names out of your mouth should be a death knell on its own, but the whole “pretending their already-expired three-month trial contract is still in effect for the full year” is a great way to find yourself pulling a Sam Bankman-Fried, except that you don’t have a side company to pull $$$ from to cover your tracks
Al sales startup AI start-up claims…
much better :3
random guess, but: "11x” is the name of the company, that’s not “eleven times”
some video-shaped AI slop mysteriously appears in the place where marketing for Ark: Survival Evolved’s upcoming Aquatica DLC would otherwise be at GDC, to wide community backlash. Nathan Grayson reports on aftermath.site about how everyone who could be responsible for this decision is pointing fingers away from themselves
[hacker voice] i’m in
it’s the same playbook, to be sure
it’s cool that you discovered a word that lets you call Asahi Lina unnecessarily dramatic and attention-seeking in a way that lets you believe you “never gave an oppinion on the matter” and that you’re just neutrally observing a scientifically studied phenomenon!
wait no “cool” isn’t the right word now is it
women do not owe you the details about why they leave a community and if you/the community as a whole feel entitled to the answer, then you have the fucking answer
i’ve heard it said before from people better at wording it than i am, but seeing this: it’s crystalizing for me that people really do see “a trans woman” as “a woman i’m still allowed to abuse”. i can call her mannish, i can tell everyone she’s making it all up, i can call her hysterical and dramatic, i can freely speculate on her mental state to the approval of my peers, and no matter what she does – leave loudly, leave quietly, stay and suffer the torment – it will always be her fault and she will always be doing it wrong
i can admit it’s possible i’m being overly cynical here and it is just sloppy journalism on Raffaele Huang/his editor/the WSJ’s part. but i still think that it’s a little suspect on the grounds that we have no idea how many times they had to restart training due to the model borking, other experiments and hidden costs, even before things like the necessary capex (which goes unmentioned in the original paper – though they note using a 2048-GPU cluster of H800’s that would put them down around $40m). i’m thinking in the mode of “the whitepaper exists to serve the company’s bottom line”
btw announcing my new V7 model that i trained for the $0.26 i found on the street just to watch the stock markets burn
That’s the opposite of what I’m saying. Deepseek is the one under scrutiny, yet they are the only one to publish source code and training procedures of their model.
this has absolutely fuck all to do with anything i’ve said in the slightest, but i guess you gotta toss in the talking points somewhere
e: it’s also trivially disprovable, but i don’t care if it’s actually true, i only care about headlines negative about AI
“the media sucks at factchecking DeepSeek’s claims” is… an interesting attempt at refuting the idea that DeepSeek’s claims aren’t entirely factual. beyond that, intentionally presenting true statements that lead to false impressions is a kind of dishonesty regardless. if you mean to argue that DeepSeek wasn’t being underhanded at all and just very innocently presented their figures without proper context (that just so happened to spurn a media frenzy in their favor)… then i have a bridge to sell you.
besides that, OpenAI is very demonstrably pissing away at least that much money every time they add one to the number at the end of their slop generator
consider this paragraph from the Wall Street Journal:
DeepSeek said training one of its latest models cost $5.6 million, compared with the $100 million to $1 billion range cited last year by Dario Amodei, chief executive of the AI developer Anthropic, as the cost of building a model.
you’re arguing to me that they technically didn’t lie – but it’s pretty clear that some people walked away with a false impression of the cost of their product relative to their competitors’ products, and they financially benefitted from people believing in this false impression.
i think you’re missing the point that “Deepseek was made for only $6M” has been the trending headline for the past while, with the specific point of comparison being the massive costs of developing ChatGPT, Copilot, Gemini, et al.
to stretch your metaphor, it’s like someone rolling up with their car, claiming it only costs $20 (unlike all the other cars that cost $20,000), when come to find out that number is just how much it costs to fill the gas tank up once
this one is a joke, i think. he is definitely on the fashy bullshit though